This forum is disabled, please visit https://forum.opencv.org
 | 1 | initial version |
Quoting the previous answer:
Keypoints are detected by Scale-Space extrema detection using Difference of Gaussian images. So the size is basically at what scale those points are detected.
This is somewhat incorrect:
The size of the patch being described is linear dependent on the keypoint scale, i.e. size = scale * magnifier. How you choose you magnifier depends on the application.
| | 2 | No.2 Revision |
Quoting the previous answer:
Keypoints are detected by Scale-Space extrema detection using Difference of Gaussian images. So the size is basically at what scale those points are detected.
This is somewhat incorrect:
The size of the patch being described is linear dependent on the keypoint scale, i.e. size = scale * magnifier. How you choose you magnifier the magnification factor depends on the application.
